version 1.92, 2013/12/31 19:40:20 |
version 1.95, 2014/01/02 22:19:41 |
|
|
/* $Id$ */ |
/* $Id$ */ |
/* |
/* |
* Copyright (c) 2011, 2012 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2011, 2012 Kristaps Dzonsons <kristaps@bsd.lv> |
* Copyright (c) 2011, 2012, 2013 Ingo Schwarze <schwarze@openbsd.org> |
* Copyright (c) 2011, 2012, 2013, 2014 Ingo Schwarze <schwarze@openbsd.org> |
* |
* |
* Permission to use, copy, modify, and distribute this software for any |
* Permission to use, copy, modify, and distribute this software for any |
* purpose with or without fee is hereby granted, provided that the above |
* purpose with or without fee is hereby granted, provided that the above |
Line 143 static void *hash_alloc(size_t, void *); |
|
Line 143 static void *hash_alloc(size_t, void *); |
|
static void hash_free(void *, size_t, void *); |
static void hash_free(void *, size_t, void *); |
static void *hash_halloc(size_t, void *); |
static void *hash_halloc(size_t, void *); |
static void mlink_add(struct mlink *, const struct stat *); |
static void mlink_add(struct mlink *, const struct stat *); |
|
static int mlink_check(struct mpage *, struct mlink *); |
static void mlink_free(struct mlink *); |
static void mlink_free(struct mlink *); |
static void mlinks_undupe(struct mpage *); |
static void mlinks_undupe(struct mpage *); |
static void mpages_free(void); |
static void mpages_free(void); |
Line 175 static int use_all; /* use all found files */ |
|
Line 176 static int use_all; /* use all found files */ |
|
static int nodb; /* no database changes */ |
static int nodb; /* no database changes */ |
static int verb; /* print what we're doing */ |
static int verb; /* print what we're doing */ |
static int warnings; /* warn about crap */ |
static int warnings; /* warn about crap */ |
|
static int write_utf8; /* write UTF-8 output; else ASCII */ |
static int exitcode; /* to be returned by main */ |
static int exitcode; /* to be returned by main */ |
static enum op op; /* operational mode */ |
static enum op op; /* operational mode */ |
static char basedir[PATH_MAX]; /* current base directory */ |
static char basedir[PATH_MAX]; /* current base directory */ |
Line 350 main(int argc, char *argv[]) |
|
Line 352 main(int argc, char *argv[]) |
|
path_arg = NULL; |
path_arg = NULL; |
op = OP_DEFAULT; |
op = OP_DEFAULT; |
|
|
while (-1 != (ch = getopt(argc, argv, "aC:d:ntu:vW"))) |
while (-1 != (ch = getopt(argc, argv, "aC:d:nT:tu:vW"))) |
switch (ch) { |
switch (ch) { |
case ('a'): |
case ('a'): |
use_all = 1; |
use_all = 1; |
Line 368 main(int argc, char *argv[]) |
|
Line 370 main(int argc, char *argv[]) |
|
case ('n'): |
case ('n'): |
nodb = 1; |
nodb = 1; |
break; |
break; |
|
case ('T'): |
|
if (strcmp(optarg, "utf8")) { |
|
fprintf(stderr, "-T%s: Unsupported " |
|
"output format\n", optarg); |
|
goto usage; |
|
} |
|
write_utf8 = 1; |
|
break; |
case ('t'): |
case ('t'): |
CHECKOP(op, ch); |
CHECKOP(op, ch); |
dup2(STDOUT_FILENO, STDERR_FILENO); |
dup2(STDOUT_FILENO, STDERR_FILENO); |
|
|
ohash_delete(&mlinks); |
ohash_delete(&mlinks); |
return(exitcode); |
return(exitcode); |
usage: |
usage: |
fprintf(stderr, "usage: %s [-anvW] [-C file]\n" |
fprintf(stderr, "usage: %s [-anvW] [-C file] [-Tutf8]\n" |
" %s [-anvW] dir ...\n" |
" %s [-anvW] [-Tutf8] dir ...\n" |
" %s [-nvW] -d dir [file ...]\n" |
" %s [-nvW] [-Tutf8] -d dir [file ...]\n" |
" %s [-nvW] -u dir [file ...]\n" |
" %s [-nvW] -u dir [file ...]\n" |
" %s -t file ...\n", |
" %s -t file ...\n", |
progname, progname, progname, |
progname, progname, progname, |
|
|
FTSENT *ff; |
FTSENT *ff; |
struct mlink *mlink; |
struct mlink *mlink; |
int dform; |
int dform; |
char *fsec; |
char *dsec, *arch, *fsec, *cp; |
const char *dsec, *arch, *cp, *path; |
const char *path; |
const char *argv[2]; |
const char *argv[2]; |
|
|
argv[0] = "."; |
argv[0] = "."; |
|
|
continue; |
continue; |
} else |
} else |
fsec[-1] = '\0'; |
fsec[-1] = '\0'; |
|
|
mlink = mandoc_calloc(1, sizeof(struct mlink)); |
mlink = mandoc_calloc(1, sizeof(struct mlink)); |
strlcpy(mlink->file, path, sizeof(mlink->file)); |
strlcpy(mlink->file, path, sizeof(mlink->file)); |
mlink->dform = dform; |
mlink->dform = dform; |
if (NULL != dsec) |
mlink->dsec = dsec; |
mlink->dsec = mandoc_strdup(dsec); |
mlink->arch = arch; |
if (NULL != arch) |
mlink->name = ff->fts_name; |
mlink->arch = mandoc_strdup(arch); |
mlink->fsec = fsec; |
mlink->name = mandoc_strdup(ff->fts_name); |
|
if (NULL != fsec) |
|
mlink->fsec = mandoc_strdup(fsec); |
|
mlink_add(mlink, ff->fts_statp); |
mlink_add(mlink, ff->fts_statp); |
continue; |
continue; |
} else if (FTS_D != ff->fts_info && |
} else if (FTS_D != ff->fts_info && |
|
|
* Try to infer this from the name. |
* Try to infer this from the name. |
* If we're not in use_all, enforce it. |
* If we're not in use_all, enforce it. |
*/ |
*/ |
dsec = NULL; |
|
dform = FORM_NONE; |
|
cp = ff->fts_name; |
cp = ff->fts_name; |
if (FTS_DP == ff->fts_info) |
if (FTS_DP == ff->fts_info) |
break; |
break; |
|
|
} else if (0 == strncmp(cp, "cat", 3)) { |
} else if (0 == strncmp(cp, "cat", 3)) { |
dform = FORM_CAT; |
dform = FORM_CAT; |
dsec = cp + 3; |
dsec = cp + 3; |
|
} else { |
|
dform = FORM_NONE; |
|
dsec = NULL; |
} |
} |
|
|
if (NULL != dsec || use_all) |
if (NULL != dsec || use_all) |
|
|
* Possibly our architecture. |
* Possibly our architecture. |
* If we're descending, keep tabs on it. |
* If we're descending, keep tabs on it. |
*/ |
*/ |
arch = NULL; |
|
if (FTS_DP != ff->fts_info && NULL != dsec) |
if (FTS_DP != ff->fts_info && NULL != dsec) |
arch = ff->fts_name; |
arch = ff->fts_name; |
|
else |
|
arch = NULL; |
break; |
break; |
default: |
default: |
if (FTS_DP == ff->fts_info || use_all) |
if (FTS_DP == ff->fts_info || use_all) |
Line 718 filescan(const char *file) |
|
Line 728 filescan(const char *file) |
|
*p++ = '\0'; |
*p++ = '\0'; |
if (0 == strncmp(start, "man", 3)) { |
if (0 == strncmp(start, "man", 3)) { |
mlink->dform = FORM_SRC; |
mlink->dform = FORM_SRC; |
mlink->dsec = mandoc_strdup(start + 3); |
mlink->dsec = start + 3; |
} else if (0 == strncmp(start, "cat", 3)) { |
} else if (0 == strncmp(start, "cat", 3)) { |
mlink->dform = FORM_CAT; |
mlink->dform = FORM_CAT; |
mlink->dsec = mandoc_strdup(start + 3); |
mlink->dsec = start + 3; |
} |
} |
|
|
start = p; |
start = p; |
if (NULL != mlink->dsec && NULL != (p = strchr(start, '/'))) { |
if (NULL != mlink->dsec && NULL != (p = strchr(start, '/'))) { |
*p++ = '\0'; |
*p++ = '\0'; |
mlink->arch = mandoc_strdup(start); |
mlink->arch = start; |
start = p; |
start = p; |
} |
} |
} |
} |
Line 742 filescan(const char *file) |
|
Line 752 filescan(const char *file) |
|
|
|
if ('.' == *p) { |
if ('.' == *p) { |
*p++ = '\0'; |
*p++ = '\0'; |
mlink->fsec = mandoc_strdup(p); |
mlink->fsec = p; |
} |
} |
|
|
/* |
/* |
Line 754 filescan(const char *file) |
|
Line 764 filescan(const char *file) |
|
mlink->name = p + 1; |
mlink->name = p + 1; |
*p = '\0'; |
*p = '\0'; |
} |
} |
mlink->name = mandoc_strdup(mlink->name); |
|
|
|
mlink_add(mlink, &st); |
mlink_add(mlink, &st); |
} |
} |
|
|
Line 768 mlink_add(struct mlink *mlink, const struct stat *st) |
|
Line 776 mlink_add(struct mlink *mlink, const struct stat *st) |
|
|
|
assert(NULL != mlink->file); |
assert(NULL != mlink->file); |
|
|
if (NULL == mlink->dsec) |
mlink->dsec = mandoc_strdup(mlink->dsec ? mlink->dsec : ""); |
mlink->dsec = mandoc_strdup(""); |
mlink->arch = mandoc_strdup(mlink->arch ? mlink->arch : ""); |
if (NULL == mlink->arch) |
mlink->name = mandoc_strdup(mlink->name ? mlink->name : ""); |
mlink->arch = mandoc_strdup(""); |
mlink->fsec = mandoc_strdup(mlink->fsec ? mlink->fsec : ""); |
if (NULL == mlink->name) |
|
mlink->name = mandoc_strdup(""); |
|
if (NULL == mlink->fsec) |
|
mlink->fsec = mandoc_strdup(""); |
|
|
|
if ('0' == *mlink->fsec) { |
if ('0' == *mlink->fsec) { |
free(mlink->fsec); |
free(mlink->fsec); |
|
|
} |
} |
} |
} |
|
|
|
static int |
|
mlink_check(struct mpage *mpage, struct mlink *mlink) |
|
{ |
|
int match; |
|
|
|
match = 1; |
|
|
|
/* |
|
* Check whether the manual section given in a file |
|
* agrees with the directory where the file is located. |
|
* Some manuals have suffixes like (3p) on their |
|
* section number either inside the file or in the |
|
* directory name, some are linked into more than one |
|
* section, like encrypt(1) = makekey(8). |
|
*/ |
|
|
|
if (FORM_SRC == mpage->form && |
|
strcasecmp(mpage->sec, mlink->dsec)) { |
|
match = 0; |
|
say(mlink->file, "Section \"%s\" manual in %s directory", |
|
mpage->sec, mlink->dsec); |
|
} |
|
|
|
/* |
|
* Manual page directories exist for each kernel |
|
* architecture as returned by machine(1). |
|
* However, many manuals only depend on the |
|
* application architecture as returned by arch(1). |
|
* For example, some (2/ARM) manuals are shared |
|
* across the "armish" and "zaurus" kernel |
|
* architectures. |
|
* A few manuals are even shared across completely |
|
* different architectures, for example fdformat(1) |
|
* on amd64, i386, sparc, and sparc64. |
|
*/ |
|
|
|
if (strcasecmp(mpage->arch, mlink->arch)) { |
|
match = 0; |
|
say(mlink->file, "Architecture \"%s\" manual in " |
|
"\"%s\" directory", mpage->arch, mlink->arch); |
|
} |
|
|
|
if (strcasecmp(mpage->title, mlink->name)) |
|
match = 0; |
|
|
|
return(match); |
|
} |
|
|
/* |
/* |
* Run through the files in the global vector "mpages" |
* Run through the files in the global vector "mpages" |
* and add them to the database specified in "basedir". |
* and add them to the database specified in "basedir". |
Line 898 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
Line 950 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
struct ohash title_table; |
struct ohash title_table; |
struct ohash_info title_info, str_info; |
struct ohash_info title_info, str_info; |
struct mpage *mpage; |
struct mpage *mpage; |
|
struct mlink *mlink; |
struct mdoc *mdoc; |
struct mdoc *mdoc; |
struct man *man; |
struct man *man; |
struct title *title_entry; |
struct title *title_entry; |
Line 932 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
Line 985 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
mparse_reset(mp); |
mparse_reset(mp); |
mdoc = NULL; |
mdoc = NULL; |
man = NULL; |
man = NULL; |
match = 1; |
|
|
|
/* |
/* |
* Try interpreting the file as mdoc(7) or man(7) |
* Try interpreting the file as mdoc(7) or man(7) |
Line 973 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
Line 1025 mpages_merge(struct mchars *mc, struct mparse *mp, int |
|
mandoc_strdup(mpage->mlinks->name); |
mandoc_strdup(mpage->mlinks->name); |
} |
} |
|
|
/* |
for (mlink = mpage->mlinks; mlink; mlink = mlink->next) |
* Check whether the manual section given in a file |
putkey(mpage, mlink->name, TYPE_Nm); |
* agrees with the directory where the file is located. |
|
* Some manuals have suffixes like (3p) on their |
|
* section number either inside the file or in the |
|
* directory name, some are linked into more than one |
|
* section, like encrypt(1) = makekey(8). Do not skip |
|
* manuals for such reasons. |
|
*/ |
|
if (warnings && !use_all && FORM_SRC == mpage->form && |
|
strcasecmp(mpage->sec, mpage->mlinks->dsec)) { |
|
match = 0; |
|
say(mpage->mlinks->file, "Section \"%s\" " |
|
"manual in %s directory", |
|
mpage->sec, mpage->mlinks->dsec); |
|
} |
|
|
|
/* |
if (warnings && !use_all) { |
* Manual page directories exist for each kernel |
|
* architecture as returned by machine(1). |
|
* However, many manuals only depend on the |
|
* application architecture as returned by arch(1). |
|
* For example, some (2/ARM) manuals are shared |
|
* across the "armish" and "zaurus" kernel |
|
* architectures. |
|
* A few manuals are even shared across completely |
|
* different architectures, for example fdformat(1) |
|
* on amd64, i386, sparc, and sparc64. |
|
* Thus, warn about architecture mismatches, |
|
* but don't skip manuals for this reason. |
|
*/ |
|
if (warnings && !use_all && |
|
strcasecmp(mpage->arch, mpage->mlinks->arch)) { |
|
match = 0; |
match = 0; |
say(mpage->mlinks->file, "Architecture \"%s\" " |
for (mlink = mpage->mlinks; mlink; |
"manual in \"%s\" directory", |
mlink = mlink->next) |
mpage->arch, mpage->mlinks->arch); |
if (mlink_check(mpage, mlink)) |
} |
match = 1; |
if (warnings && !use_all && |
} else |
strcasecmp(mpage->title, mpage->mlinks->name)) |
match = 1; |
match = 0; |
|
|
|
putkey(mpage, mpage->mlinks->name, TYPE_Nm); |
|
|
|
if (NULL != mdoc) { |
if (NULL != mdoc) { |
if (NULL != (cp = mdoc_meta(mdoc)->name)) |
if (NULL != (cp = mdoc_meta(mdoc)->name)) |
putkey(mpage, cp, TYPE_Nm); |
putkey(mpage, cp, TYPE_Nm); |
Line 1711 utf8key(struct mchars *mc, struct str *key) |
|
Line 1731 utf8key(struct mchars *mc, struct str *key) |
|
* Parse the escape sequence and see if it's a |
* Parse the escape sequence and see if it's a |
* predefined character or special character. |
* predefined character or special character. |
*/ |
*/ |
|
|
esc = mandoc_escape |
esc = mandoc_escape |
((const char **)&val, &seq, &len); |
((const char **)&val, &seq, &len); |
if (ESCAPE_ERROR == esc) |
if (ESCAPE_ERROR == esc) |
break; |
break; |
|
|
if (ESCAPE_SPECIAL != esc) |
if (ESCAPE_SPECIAL != esc) |
continue; |
continue; |
if (0 == (u = mchars_spec2cp(mc, seq, len))) |
|
continue; |
|
|
|
/* |
/* |
* If we have a Unicode codepoint, try to convert that |
* Render the special character |
* to a UTF-8 byte string. |
* as either UTF-8 or ASCII. |
*/ |
*/ |
cpp = utfbuf; |
|
if (0 == (sz = utf8(u, utfbuf))) |
|
continue; |
|
|
|
|
if (write_utf8) { |
|
if (0 == (u = mchars_spec2cp(mc, seq, len))) |
|
continue; |
|
cpp = utfbuf; |
|
if (0 == (sz = utf8(u, utfbuf))) |
|
continue; |
|
sz = strlen(cpp); |
|
} else { |
|
cpp = mchars_spec2str(mc, seq, len, &sz); |
|
if (NULL == cpp) |
|
continue; |
|
if (ASCII_NBRSP == *cpp) { |
|
cpp = " "; |
|
sz = 1; |
|
} |
|
} |
|
|
/* Copy the rendered glyph into the stream. */ |
/* Copy the rendered glyph into the stream. */ |
|
|
sz = strlen(cpp); |
|
bsz += sz; |
bsz += sz; |
|
|
buf = mandoc_realloc(buf, bsz); |
buf = mandoc_realloc(buf, bsz); |
|
|
memcpy(&buf[pos], cpp, sz); |
memcpy(&buf[pos], cpp, sz); |
pos += sz; |
pos += sz; |
} |
} |